LAUSR

Background Healthcare disparity related to sociodemographic, environmental and community factors is well recognized as a cause of worse health outcomes. Its importance in intensive procedures such as allogeneic hematopoietic cell transplantation (alloHCT) is unknown. The County Health Rankings is publicly available data based on zip codes, updated annually and measuring 35+ health factors and health outcomes that can be used to calculate community risk scores. Higher scores indicate less healthy communities. We tested whether community risk scores of individual patients and of transplant centers (TC) are associated with post alloHCT outcomes, adjusting for other known disease and treatment factors. Methods Data were provided by the Center for International Blood and Marrow Transplant Research (CIBMTR) for all US first alloHCT for adult recipients with malignant diseases performed from 2014 to 2016. Patient and TC community risk scores were derived by normalizing County Health Rankings and calculated based on patients' zip codes of residence and TC zip codes, respectively. Multivariable analysis using Cox proportional hazards regression was conducted, and associations were tested between the community risk scores and overall survival (OS), relapse, transplant-related mortality (TRM), and death after relapse. Results A total of 17,812 alloHCT cases from 170 US TCs were included in the main analysis. Most recipients were white (85%) and received grafts from a matched unrelated donor (42%), HLA-identical sibling (29%) or a mismatched related donor (11%). Acute myelogenous leukemia (42%), acute lymphoblastic leukemia (14%), myelodysplastic syndromes (MDS) (17%) and non-Hodgkin lymphoma (11%) were the most common indications for HCT. The median age was 56 years (range 18-83). Median follow up was 24 months. The median community risk score was -0.21 (range, -1.37 to 2.10; standard deviation (SD)=0.42). Higher patient community risk scores were associated with inferior overall survival (for one SD increase in patient community risk score, HR for mortality is 1.04, 95% CI 1.01-1.07, p=0.008). TC community risk scores were not significantly associated with OS. On preliminary analysis, patient community risk scores were not significantly associated with relapse (HR 0.98, 95% CI 0.95-1.01, p=0.11) or TRM (HR 0.98, 95% CI 0.95-1.00, p=0.068). However, higher patient community risk scores were associated with increased risk of post relapse mortality (HR 1.05, 95% CI 1.02-1.08, p=0.001). TC community risk scores were not significantly associated with any of the outcomes. Conclusion Patients from counties with worse community health have lower overall survival after alloHCT. Additional studies are needed to understand why these patients have shorter survival after relapse.